Clinical significance of diastolic wall strain in heart failure with preserved ejection fraction: A cross-sectional observational analysis

Abstract

Background

Diastolic wall strain (DWS) is based on linear elastic theory, which shows that impaired diastolic wall thinning reflects resistance to deformation in diastole and thus, increased diastolic myocardial stiffness. We aim to explore the role of DWS in patients with heart failure with preserved ejection fraction (HFpEF) in terms of correlation with indices of HFpEF.

Methods

Study enrolled 53 patients with exertional dyspnoea and normal left ventricular ejection fraction. Forty patients fulfilled the criteria for HFpEF according to ESC 2023 criteria.

Results

Two groups were analysed – Group 1 with criteria of HFpEF fulfilled and Group 2 with those who did not. Echocardiographic indices including relative wall thickness (RWT), left ventricular mass index (LVMI), E/e’, left atrial volume index (LAVI) and DWS were numerically different on comparison with group 2, with LVMI, LAVI and E/e’ statistically significant. Also mean Global Longitudinal Strain (GLS) was found to be −13.05%. Group 1 was divided into HFpEF with DWS ≤ median and HFpEF with DWS > median. Echocardiographic indices showed statistically higher LVMI and atrial filling fraction. This finding showed that patients with reduced DWS were more likely to have diastolic dysfunction. Also, it was found that DWS had a statistically significant correlation with LVMI, LAVI and RWT. Lower DWS had abnormal GLS. Limitations include small sample size.

Conclusion

Although difference in DWS between HFpEF and controls did not reach statistical significance, stratification by median value showed significant correlation of DWS with myocardial relaxation parameters. Also, with significant correlation with increased atrial filling fraction higher N-terminal pro-B type natriuretic peptide and correlation with impaired GLS, our study supports DWS as a potential research tool in evaluation of HFpEF.

Graphical abstract.

This is a visual representation of the abstract.

Background

Heart failure with preserved ejection fraction (HFpEF) is a complex clinical syndrome that makes up around 56% of cases of heart failure and its prevalence is on the rise. ¹ Indian data from the National Heart Failure Registry, which had enrolled 10,851 acute HF patients showed 12.7% prevalence of HFpEF among Indian population. ² Though, HFpEF and heart failure with reduced ejection fraction (HFrEF) exhibit similar outcome in terms of hospitalisation rate, hospitalisation duration and impact on quality of life, patients with HFpEF are currently recognised as having the greatest unmet need in cardiology, with changing paradigms in diagnosis, therapeutic options that currently caters only to alleviate symptoms and with scarce data on outcomes of such patients. ³ LV diastolic dysfunction (LVDD) plays a fundamental role in the pathophysiology of HFpEF and indices of diastolic dysfunction, measured by echocardiography, is an important tool in diagnosis. ⁴ LVDD is defined by impairment in relaxation and an increase in viscoelastic chamber stiffness, that causes elevated ventricular filling pressures, leading to symptomatic heart failure.^5,6 Diastolic stiffness inferred from Doppler indices, reflect filling pressures and myocardial relaxation without directly measuring the diastolic stiffness. ⁷ Diastolic wall strain (DWS) is based on the linear elastic theory. According to which, impaired diastolic wall thinning reflects the resistance to deformation in diastole, contributing to diastolic myocardial stiffness. Also, invasive animal model experiments have clearly shown correlation of DWS with the diastolic stiffness constant. ⁸ Previous studies showed conflicting results of correlation between DWS and other echocardiographic parameters of diastolic dysfunction. Hence this study aims to explore the role of DWS in patients with HFpEF in terms of correlation with indices of HFpEF.

Materials and methods

This cross sectional observational study enrolled 53 consecutive ambulatory patients of more than 18 years of age, with symptoms of heart failure (dyspnoea NYHA II–IV) and normal left ventricular ejection fraction (LVEF), presenting to Institute of Cardiology, RGGGH and MMC, Chennai, India. Written informed consent was obtained from all patients enrolled. Patients with acute myocardial infarction on admission, patients with severe aortic stenosis, aortic regurgitation, mitral stenosis or mitral regurgitation due to structural changes in valves detected using TTE on admission, any prior echocardiographic measurement of LVEF <40%, probable alternative diagnoses that could account for the patient's HF symptoms (i.e. dyspnoea, fatigue), such as significant pulmonary disease (patient reported history of bronchial asthma/obstructive lung disease, old pulmonary tuberculosis prolonged history of expectorant cough), anaemia (haemoglobin in male less than 13 g/dL and in females less than 12 g/dL) were excluded. Of those 53 patients, 40 patients fulfilled the criteria for HFpEF according to European Society of Cardiology 2023 criteria (which was based on elevated N-terminal pro-B type natriuretic peptide (NT pro-BNP) cut-off). ⁹ Patients who fulfilled the criteria for HFpEF (n = 40) were labelled as group 1, and those who did not fulfil (n = 13) were labelled as Group 2.

Data on clinical history including past history of diabetes, hypertension, chronic kidney disease, laboratory parameters including complete blood count, renal function tests, NT pro-BNP, echocardiographic indices at baseline were collected from all patients included in the study. Echo was done using Philips Affiniti 70 elevate by a single operator. Images were recorded in the standard views: parasternal long axis, parasternal short axis, at the base, at the level of the papillary muscle and the apex and in apical four-chamber and apical two-chamber views. Tissue Doppler and trans-mitral Doppler indices were measured in standard views. Left ventricular mass index (LVMI) was estimated according to the Devereux formula indexed to body surface area. ¹⁰ Left atrial volume was calculated using ellipsoid method (D1 × D2 × D3 × 0.523) and indexed to body surface area (left atrial volume index (LAVI)). Atrial filling fraction was calculated for those with HFpEF, using the formula (A wave VTI/(E.VTI + A.VTI)) and DWS using the following formula: DWS = (Posterior wall thickness in systole (PWs) – Posterior wall thickness in diastole (PWd))/PWs. PWs was measured in end systole and PWd measured in end diastolic phase of cardiac cycle and average of three values were taken to reduce the intra-observer variability. Global Longitudinal Strain (GLS) was derived from three apical echocardiographic views (4-chamber, 3-chamber and 2-chamber).

Statistical analysis was made using PSPP version 1.6.2. Analysis for categorical and continuous variables was done by ANOVA. Correlation between DWS, e’, E/e’, LVMI, relative wall thickness (RWT) and LAVI were done using Pearson's co-efficient. The results with p-values <.05 were considered statistically significant.

Results

Baseline characteristics of Group 1 and Group 2 were given in Table 1. Mean age of Group 1 patients was 56.25 years with predominant females (62.5%). Diabetes and hypertension were found in 40% and 52.5% of patients respectively, with prevalence of both diabetes and hypertension higher in Group 1. Average BMI was 25.68 kg/m². Mean NT-Pro BNP was 1988.78 pg/mL. Echocardiographic indices included mean RWT of 0.44, mean LVMI of 115.14 g/m², mean E/e’ of 16.49 and a mean DWS of 0.25, which were noticeably different on comparison with Group 2, supporting the importance of these parameters in LVDD and in diagnosis of HFpEF. Of the echocardiographic parameters, DWS was lower in Group 1, though statistically not significant. Mean LAVI in Group 1 was 20.76 mL/m², which was significantly higher compared to Group 2 (p = .017). Other parameters which were statistically significant were LVMI (p = .028) and E/e’ (p = .01). HFA-PEFF score was 6 in 20% (n = 8), 5 in 50% (n = 20), 4 in 22.5% (n = 9) and 3 in 7.5% (n = 3), of those in Group 1. On comparison with Group 2, HFA-PEFF score was 2 in majority (61.5%), with a score of 3 in 23%. Also mean GLS in those with HFpEF was found to be −13.05%, which was well below the normal limit. Prevalence of atrial fibrillation was higher in patients with HFpEF, as expected. Etiology among those with HFpEF was predominantly ischaemic heart disease (n = 28), of which 22 patients underwent coronary angiogram, which showed significant double vessel disease involving predominantly left anterior descending artery in eight patients, significant triple vessel disease in six patients (with two patients showing left main involvement), six patients showing single vessel disease of left anterior descending artery and two patients with slow flow coronaries. Of the remaining 12 patients, one patient had cardiac amyloidosis, five patients showed restrictive physiology. Etiology could not be elucidated in the remaining.

Table 1.

Comparison of baseline characteristics between those fulfilling HFpEF criteria (Group 1) and those not fulfilling HFpEF criteria (Group 2).

Parameters	Group 1 (n = 40) Mean (SD)	Group 2 (n = 13) Mean (SD)	p-value
Age (years)	56.25 (14.28)	57.38 (8.18)	.787
Sex	Male – 15 (37.5%)	Male – 5 (38.46%)	.074
	Female – 25 (62.5%)	Female – 8 (61.53%)
Diabetes (No. of patients)	16 (40%)	2 (15.38%)	.02
Hypertension (No. of patients)	21 (52.5%)	6 (46.15%)	.891
CKD (No. of patients)	4 (10%)	0	-
Body mass index (BMI) (Kg/m2)	25.68 (4.90)	26.80 (7.19)	.529
LVEF (%)	59.73 (6.71)	62.23 (3.47)	.204
NT PRO-BNP (pg/mL)	1988.78 (2314.11)	83.23 (40.25)	.005
RWT	0.44 (0.14)	0.37 (0.06)	.081
LVMI (g/m2)	115.14 (43.36)	84.35 (40.58)	.028
Average e’	5.99 (2.40)	7.23 (1.11)	.079
E/e’	16.49 (8.11)	10.37 (2.98)	.01
LAVI (mL/m2)	20.76 (10.17)	13.56 (4.32)	.017
MV deceleration time (ms)	192.13 (57.04)	181.92 (39.91)	.554
DWS	0.25 (0.09)	0.29 (0.07)	.098
GLS (%)	−13.05 (10.19)
ECG parameters: (No. of patients)
• Atrial fibrillation	4 (10%)	0
• Atrial flutter	1 (2.5%)	0

HFpEF: heart failure with preserved ejection fraction; LVEF: left ventricular ejection fraction; NT PRO-BNP: N-terminal pro-B type natriuretic peptide; RWT: relative wall thickness; LVMI: left ventricular mass index; LAVI: left atrial volume index; DWS: diastolic wall strain; GLS: Global Longitudinal Strain. Bold values showed statistical significance on analysis.

Patients of Group 1 were further divided on the basis of DWS into—those with DWS less than or equal to the median value (HFpEF with DWS < median) and those with DWS more than the median value (HFpEF with DWS > median), whose echocardiographic parameters were given in Table 2. This comparison was intended to further explore the significance of DWS as a part of exploratory subgroup analysis. Echocardiographic indices were comparable in both the groups except for LVMI which showed a statistically significant higher value in those with reduced DWS. Also, atrial filling fraction which denotes the relative contribution of atrial booster function was also significantly higher in the former group. It was found that patients with decreased DWS (less than median) had worse GLS than those with higher DWS with a statistical significance (p-value .012), with overall GLS being abnormal.

Table 2.

Echocardiographic parameters between subsets of HFpEF population showing that more reduced DWS, more abnormal the diastolic parameters (subgrouping based on median DWS and is exploratory).

Echocardiographic parameters	HFpEF with DWS less than or equal to median value (n = 22) mean (SD)	HFpEF with DWS more than median value (n = 18) mean (SD)	p-value
LVEF (%)	59.23 (7.25)	60.33 (6.14)	.610
RWT	0.48 (0.15)	0.41 (0.10)	.106
LVMI (g/m2)	128.47 (46.89)	98.833 (32.95)	.030
Average e’	5.34 (1.83)	6.79 (2.80)	.057
E/e’	15.78 (8.38)	17.36 (7.94)	.548
MV deceleration time (ms)	190.48 (51.79)	194.06 (64.10)	.848
DWS	0.19 (0.05)	0.33 (0.05)	<.001
Atrial filling fraction	0.36 (0.11)	0.25 (0.15)	.013
GLS (%)	−10.69 (3.98)	−13.39 (3.99)	.012

HFpEF: heart failure with preserved ejection fraction; LVEF: left ventricular ejection fraction; RWT: relative wall thickness; LVMI: left ventricular mass index; DWS: diastolic wall strain; GLS: Global Longitudinal Strain. Bold values showed statistical significance on analysis.

On further correlation of DWS with other standard echocardiographic parameters, it was found that DWS correlated with statistical significance with LVMI, LAVI and RWT. Tissue Doppler indices did not show a statistically significant correlation with DWS. Correlation was assessed by Pearson's co-efficient, whose parameters are given in Table 3. DWS showed a significant correlation with NT pro-BNP with a p-value of .030.

Table 3.

Correlation of DWS with variables predicting diastolic dysfunction (as DWS becomes reduced, parameters that showed increased values had a negative Pearson coefficient).

Parameters	p-value (<.05 – significant)	Pearson's R co-efficient
RWT	.023	−0.317
LVMI	.010	−0.364
LAVI	.048	−0.266
Average e’	.080	0.227
E/e’	.201	0.136
NT Pro-BNP	.030	−0.309
GLS	.169	0.104

NT PRO-BNP: N-terminal pro-B type natriuretic peptide; RWT: relative wall thickness; LVMI: left ventricular mass index; LAVI: left atrial volume index; DWS: diastolic wall strain; GLS: Global Longitudinal Strain. Bold values showed statistical significance on analysis.

Discussion

When baseline characteristics were compared between those with HFpEF and those without HFpEF, diabetes was one risk factor which was statistically prevalent in the former, reinforcing its role in pathogenesis of HFpEF. In our cohort, it was confirmed that on average, DWS was lower in patients with HFpEF. DWS is a simple, non-invasive index calculated from standard echocardiographic parameters. Lower DWS was suggestive of increased myocardial stiffness. The mean DWS in HFpEF patients in our study was 0.25 ± 0.09, which was lower than that described by Ohtani et al. ¹¹ and was comparable to those described by Spinarova et al. ¹² Although, DWS was lower in patients with HFpEF, the difference did not reach statistical significance. This finding was contradictory from previous studies, which showed DWS being a significant marker. Significance of DWS, thus required further investigation. Hence patients with HFpEF were further divided into two groups, to explore whether lower value of DWS had any significance in the diagnosis of LVDD.

It was found that patients with HFpEF with lower DWS (< median value) showed more abnormal geometry in terms of higher LVMI and RWT, with LVMI showing statistical significance. These findings were comparable to previous study findings as elaborated by Ohtani et al. ¹¹ Also, atrial filling fraction was significantly higher in patients with lower DWS. In diastolic dysfunction, the atrial filling fraction is typically increased due to the impaired ability of the left ventricle to accept blood during diastole. Atrial filling fraction simply measures the contribution of atrium in diastolic filling, with increased contribution needed in situations where the ventricle has impaired relaxation. Correlation of lower DWS with atrial filling fraction (p-value of .013), in our study further adds on to the value of DWS as a parameter in reflecting increased myocardial stiffness and in-turn contributing its role in diagnosing diastolic dysfunction. This further strengthens the fact that lower the DWS, more significant is the diastolic dysfunction in HFpEF. But there was an inverse difference in tissue Doppler indices between those with lower DWS and higher DWS, with higher value of E/e’ in the latter group, which might indicate the limitation of tissue Doppler indices in assessing complex myocardial deformation patterns due to its reliance on a single velocity measurement. ¹³ It was found that LAVI had a significant correlation with DWS as seen in Table 3, which was done on exploratory basis consistent with previous studies. ¹²

Our study showed the presence of abnormal GLS in patients with HFpEF, reiterating the observations from previous studies. This observational analysis, though only supportive and not of pathophysiological significance, puts forth the question of whether impaired GLS in HFpEF portends future overt ventricular systolic dysfunction, which needs further specific research work in this regard In such a scenario, routine assessment of GLS in patients with HFpEF might become important. There are previous studies on correlation of DWS with systolic indices, particularly of speckle tracking echocardiographic parameters of strain indices. They had quoted the potential mechanism of abnormal systolic mechanics in patients with HFpEF, playing a significant role in poor patient outcomes, especially in those with lower DWS. ¹⁴ Though in our study, we could not establish a direct significant correlation between GLS and DWS (p-value .169), subgroup analysis of DWS below the median showed an abnormal GLS with reduced DWS, with a statistical significance. This might be explained by the underlying pathological process that, more advanced the myocardial fibrosis and stiffening, more impaired would be the longitudinal shortening, as measured by GLS and passive stretching, as measured by DWS. This supports the concept that abnormal myocardial mechanics strongly affects the DWS in HFpEF.

Diagnosis of HFpEF continues to evolve, and there are still lacunae in non-invasive measures of left ventricle filling pressures. HFA-PEFF diagnostic score, being considered as one of the better tools, relies mainly on tissue Doppler indices and myocardial relaxation parameters (LVMI and RWT). Because DWS shows consistent correlation with the latter parameters, it may serve as a complementary marker of adverse ventricular remodelling. Furthermore, our study showed a statistically significant correlation of DWS with NT pro BNP, further strengthening the potential the role of DWS in diagnosis of HFpEF. With previous studies having proven the association of reduced DWS with poorer outcome in patients with HFpEF, these findings open the discussion of DWS as a parameter for risk stratification of patients with HFpEF. Dedicated outcome studies focussed on DWS might strengthen this evidence.

Limitations of our study include less sample size. There was a lack of invasive gold standard measurements of diastolic stiffness. Also, there is a possibility of regional assessment of LV stiffness at the posterior wall not reflecting the global LV myocardial stiffness. Potential involvement of confounding factors like presence of hypertension, chronic kidney disease and BMI is another limitation. Inter and intra-observer variability by statistical analysis, in measurement of DWS was not addressed. Most of our patients had ischaemic heart disease as the etiology of HFpEF, which might limit generalizability to other etiologies, as LV wall thickness could be altered with presence or absence of hypertension, which in-turn affects the DWS.

Conclusion

Thus, DWS being a relatively simple echocardiographic marker of myocardial stiffness, was shown to have moderate, yet statistically significant association with LVMI, RWT and LAVI in patients with HFpEF. Although the difference in DWS between HFpEF and controls did not reach statistical significance, exploratory subgroup analysis by median value showed a significant correlation of DWS with consistent myocardial relaxation parameters. Also, with greater dependence on atrial contribution to filling (suggested by elevated atrial filling fraction) and significantly higher NT pro-BNP levels, incorporation of this novel, yet simple index of LVDD into the diagnostic armamentarium of HFpEF as an alternative marker, might be considered. Our study also establishes a crucial link between impaired GLS and reduced myocardial elasticity during diastole, reinforcing the concept of myocardial stiffness being a common pathological substrate for both systolic and diastolic dysfunction in HFpEF. As the conventional diastolic dysfunction indices are more deranged with reduced DWS, there appears a trend towards more advanced disease substrate, as the value of DWS becomes much reduced. Larger prospective studies are warranted to validate the exact diagnostic cut-off and to determine its prognostic value in HFpEF. This study provides a strong rationale for exploring further in-depth role of DWS with future large-scale, outcome driven studies.

Abbreviations

A velocity

transmitral velocity during atrial systole

ANOVA

analysis of Variance

BMI

body mass index

DWS

diastolic wall strain

E/e’

ratio of early diastolic transmitral velocity to early diastolic mitral annular velocity

GLS

global longitudinal strain

HF

heart failure

HFA-PEFF

Heart Failure Association-pre-test assessment, echocardiography and natriuretic peptide, functional testing, final etiology.

HFpEF

heart failure with preserved ejection fraction

LAVI

left atrial volume index

LV

left ventricle

LVDD

left ventricular diastolic dysfunction

LVEF

left ventricular ejection fraction

LVMI

left ventricular mass index

NHFR

National Heart Failure Registry

NT pro-BNP

N-terminal pro-B type natriuretic peptide

NYHA

New York Heart Association

QoL

quality of life

RWT

relative wall thickness

TTE

transthoracic echocardiography

VTI

velocity time integral